Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=gpcl20

Download by: [University of Kent] Date: 23 November 2015, At: 23:49

Psychology, Crime & Law

ISSN: 1068-316X (Print) 1477-2744 (Online) Journal homepage: http://www.tandfonline.com/loi/gpcl20

Comparing the psychological characteristics of un-apprehended firesetters and non-firesetters livingin the UK

Emma R. Barrowcliffe & Theresa A. Gannon

To cite this article: Emma R. Barrowcliffe & Theresa A. Gannon (2015): Comparing thepsychological characteristics of un-apprehended firesetters and non-firesetters living in the UK,Psychology, Crime & Law, DOI: 10.1080/1068316X.2015.1111365

To link to this article: http://dx.doi.org/10.1080/1068316X.2015.1111365

Accepted author version posted online: 22Oct 2015.Published online: 20 Nov 2015.

Submit your article to this journal

Article views: 2

View related articles

View Crossmark data

Comparing the psychological characteristics ofun-apprehended firesetters and non-firesetters living in the UKEmma R. Barrowcliffe and Theresa A. Gannon

Centre of Research and Education in Forensic Psychology, School of Psychology, Keynes College, University ofKent, Canterbury, Kent, UK

ABSTRACTDeliberate firesetting research predominantly focuses onapprehended populations. In contrast, this paper focuses on theprevalence and characteristics of un-apprehended firesetters livingin the UK. Social media was utilized to recruit 232 participants foran online questionnaire. Two hundred and twenty-five peopleanswered a question relating to deliberate firesetting. Fortyparticipants (17.78%) indicated that they had ignited a deliberatefire and were classified as un-apprehended firesetters. Firesettingwas most common in childhood and adolescence. Relative to non-firesetters, un-apprehended deliberate firesetters were more likelyto report; a diagnoses of a psychiatric illness, a diagnosis of abehavioural problem, having been suspended from school, ahistory of suicide attempts, experimenting with fire before the ageof 10 years old, and having a family member who also ignited adeliberate fire. Un-apprehended firesetters also scored significantlyhigher compared to non-firesetters on the Fire Setting Scale and theFire Proclivity Scale [Gannon, T. A., & Barrowcliffe, E. R. (2012).Firesetting in the general population: The development andvalidation of the Fire Setting and Fire Proclivity Scales. Legal andCriminological Psychology, 17(1), 105–122], the Fire Interest RatingScale [Murphy, G. H., & Clare, I. C. H. (1996). Analysis of motivationin people with mild learning disabilities (mental handicap) who setfires. Psychology, Crime & Law, 2(3), 153–164], the Novaco AngerScale and Provocation Inventory [Novaco, R. W. (2003). The NovacoAnger Scale and Provocation Inventory: NAS-PI. Los Angeles, CA:Western Psychological Services], the Boredom Proneness Scale –Short Form [Vodanovich, S. J., Wallace, J. C., & Kass, S. J. (2005). Aconfirmatory approach to the factor structure of the BoredomProneness Scale: Evidence for a two-factor short form. Journal ofPersonality Assessment, 85(3), 295–303], and the Measure of CriminalAttitudes and Associates Scale [Mills, J. F., & Kroner, D. G. (1999).Measures of criminal attitudes and associates: User guide.Unpublished instrument and user guide].

ARTICLE HISTORYReceived 22 June 2015Accepted 13 October 2015

KEYWORDSArson; firesetter; deliberatefiresetting; adolescent; un-apprehended

In Great Britain, 23,662 deliberate fires were ignited between April 2012 and March 2013which resulted in 56 deaths and 1225 non-fatal injuries (Department for Local

© 2015 Taylor & Francis

CONTACT Emma R. Barrowcliffe eb34@kent.ac.uk

PSYCHOLOGY, CRIME & LAW, 2015http://dx.doi.org/10.1080/1068316X.2015.1111365

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

Communities and Local Government, 2014). Arson is a legal term which refers to fires thatare deliberately ignited with the intent to destroy property (Kolko, 2002; Williams, 2005).However, the term arson is restrictive and varies across jurisdictions (Gannon & Pina,2010). Therefore in this paper the term firesetting is used to refer to all acts of deliberateignition, which may not have been legally recorded as arson. The term firesetter refers tothe perpetrator of the deliberate ignition.

Deliberate firesetting was estimated to cost the UK approximately £2.3 billion in 2008(Department for Communities and Local Government, 2011). However, relative to othercrimes arson has the poorest detection rate in England and Wales (Smith, Taylor, &Elkin, 2013). For example, out of 19,306 arson offences reported in 2013, only 2316 (or12%) resulted in official police detection (Smith et al., 2013), and 1503 people were pro-ceeded against in court (adult males = 934, adult females = 265; adolescent males = 253,adolescent females = 381; Justice Statistics Analytical Services, 2015). Of these a total of76% of the adults and 72% of the adolescents were found guilty. Therefore it is apparentand problematic that the vast majority of arson perpetrators remain un-apprehended. Inorder to manage the behaviour of deliberate firesetters it is essential to have a compre-hensive understanding of their characteristics (Doley, 2003). Unfortunately, however, com-pared to other types of offending firesetting is one of the least understood behaviours(Davis & Lauber, 1999; Dickens, Sugarman, & Gannon, 2012). To date a small amount offiresetting research has tended to focus on apprehended populations such as childrenand adolescents (Fessler, 2006; Kazdin & Kolko, 1986; Root, MacKay, Henderson, DelBove, & Warling, 2008), prisoners (Gannon et al., 2013; O’Sullivan & Kelleher, 1987; Saps-ford, Banks, & Smith, 1978), and psychiatric patients (Gannon & Pina, 2010; O’Sullivan &Kelleher, 1987; Räsänen, Hakko, & Väisänen, 1995; Tennent, McQuaid, Loughnane, &Hands, 1971; Tyler & Gannon, 2012).

These apprehended firesetters share common characteristics. For example both adoles-cent and adult apprehended firesetters tend to be Caucasian (Gannon, 2010; Koson &Dvoskin, 1982) and male (Bradford, 1982; Muller, 2008; Pettiway, 1987; Räsänen et al.,1995; Rautaheimo, 1989). Apprehended firesetters are characterized by poor developmen-tal experiences such as victimization or abuse during childhood (Gannon, 2010; Noblett &Nelson, 2001; Saunders & Awad, 1991), separation from parents (Macht & Mack, 1968;Saunders & Awad, 1991; Tennent et al., 1971), poor education (Harmon, Rosner, & Wieder-light, 1985; Lewis & Yarnell, 1951; Rautaheimo, 1989), and low IQ (Bradford, 1982; Harmonet al., 1985; Lewis & Yarnell, 1951; Rautaheimo, 1989). Many apprehended firesetters alsotend to have mental health issues (Räsänen et al., 1995; Tyler & Gannon, 2012) and sub-stance abuse problems (Jayaraman & Frazer, 2006). In addition, apprehended firesettershave a tendency to ignite fires close to home (Bradford, 1982; Rautaheimo, 1989; Wachiet al., 2007).

The list of motivations underpinning firesetting is comprehensive. For example appre-hended firesetters report igniting fires as a result of peer pressure (Molnar, Keitner, &Harwood, 1984; Swaffer & Hollin, 1995), as a form of communication (Geller, 1992), forvandalism or to create excitement (Gannon & Pina, 2010; Icove & Estepp, 1987; Inciardi,1970), and in order to conceal another crime (Dennet, 1980). Other inclinations includeself-protection (Tyler et al., 2014), political motivation (e.g. terrorist attacks, riots; Prins,1994), and self-injury or suicide (Jayaraman & Frazer, 2006; McKerracher & Dacre, 1966;Noblett & Nelson, 2001; O’Sullivan & Kelleher, 1987; Swaffer & Hollin, 1995). However,

2 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

the predominant motivation behind both adolescent and adult firesetting is revenge(Gannon, Ó Ciardha, Doley, & Alleyne, 2012; Koson & Dvoskin, 1982; Lewis & Yarnell,1951; O’Sullivan & Kelleher, 1987; Rix, 1994; Swaffer & Hollin, 1995).

Although apprehended firesetters share similar characteristics, evidence suggests thatfiresetters are a unique offending population. Gannon et al. (2013) found that male adultapprehended firesetters could be differentiated from other offenders on fire-relatedfactors (i.e. more identification with fire, interest in everyday and serious fires, attitudesaimed at legitimizing firesetting as ‘normal’, and less perceived fire safety knowledge),emotional/self-regulation factors (i.e. firesetters report significantly more anger-relatedcognitions, physiological arousal to anger, and are more susceptible to provocation),and self-concept factors (i.e. firesetters had lower levels of self-esteem). Gannon et al.(2013) conclude that relative to other non-firesetting offenders, firesetters incarceratedin prisons are a special group of offenders who hold unique psychological characteristics.

The vast majority of research has almost exclusively concentrated on the characteristicsof apprehended firesetters with little consideration for the prevalence and psychologicalcharacteristics of firesetters who remain un-apprehended. It is therefore inappropriate togeneralize these findings to all firesetting populations. Relatively little is known about fire-setters who manage to evade detection and to our knowledge there are only a few studiesconcerned with un-apprehended firesetters. The first of these studies was not specificallydesigned to assess firesetting but instead utilized nationally representative data from aNational Epidemiologic Survey on Alcohol and Related Conditions in the USA (NESARC;Blanco et al., 2010; Vaughn et al., 2010). Participants were interviewed face to face, andthose who responded positively to the question ‘in your entire life, did you ever start afire on purpose to destroy someone else’s property or just to see it burn?’ were classifiedas firesetters (1–1.13%; Blanco et al., 2010; Vaughn et al., 2010). The majority of firesettingbehaviour was reported to occur during adolescence (i.e. ≤15 years; Blanco et al., 2010).

The participants classified as firesetters (n = 407) in the NESARC study were comparedto the non-firesetters (n = 41,552). Results indicated that being male, born in the USA,increased engagement in antisocial behaviour (e.g. destroying property), having a highannual income (>$70,000), and never having married were risk factors for firesettingbehaviour (Blanco et al., 2010; Vaughn et al., 2010). Firesetting was also associated withhaving a Diagnostic and Statistical Manual of Mental Disorders 4th edition diagnosis ofantisocial personality disorder, drug dependence, bipolar disorder, and pathological gam-bling (Blanco et al., 2010).

However, the NESARC study had a number of limitations. For example, respondentswereinterviewed face to face which may have resulted in a reluctance to answer the firesettingquestion truthfully for fear of reprisals (Dickens et al., 2012; Gannon & Barrowcliffe, 2012).Dickens et al. (2012) also note that the single question assessing firesetting behaviour isextremely vague. For example the section of the question, ‘on purpose to destroysomeone else’s property or just to see it burn’may have resulted in an over representationof firesetting as some respondentsmay havemisidentified experimentationwith fire (just tosee it burn) asmeeting the criteria for starting afire. Furthermore, information relating to thetypes of fires ignited, the severity of the fire, whether the respondent was ever formallyapprehended or received any therapy for their firesetting are lacking.

Gannon and Barrowcliffe (2012) and Barrowcliffe and Gannon (2015) have recentlyconducted research assessing the prevalence and characteristics of un-apprehended

PSYCHOLOGY, CRIME & LAW 3

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

firesetters living in the UK. In both studies adult participants were provided with veryspecific criteria regarding the types of firesetting the researchers were interested in. Par-ticipants were directed to only disclose information relating to fires deliberately ignited asa result of boredom, or to create excitement, fires set as a result of peer pressure, toexpress feelings, as an act of vandalism, revenge or to conceal another crime. Participantswere directed not to report fires set before the age of 10 years,2 ignited accidentally, or aspart of organized events such as bonfires.

Participants in both studies who indicated they had ignited a deliberate fire were thenasked to report detailed information about the fire, for example the motive behindignition, the number of ignition points, and the paraphernalia used. Participants also com-pleted scales designed to specifically measure fire interest and antisocial behaviour. TheFire Setting Scale (FSS) comprises two subscales measuring fire interest and antisocialbehaviour. The Fire Proclivity Scale (FPS) contains six hypothetical firesetting scenarios(with varying degrees of severity) designed to measure firesetting proneness or proclivity.The FPS requires participants to imagine themselves as the perpetrator in each of the fire-setting scenarios, and rate their likelihood of fire fascination, behavioural propensity to actsimilarly, arousal, and general antisocialism in relation to each scenario.

In the first study conducted by Gannon and Barrowcliffe (2012), 168 participants (109females) met with the researchers to complete the research, and placed their responsesin an unlabelled envelope to protect anonymity. Of the 168 participants, 11% (n = 18)were classified as deliberate firesetters, and none reported having been apprehendedfor firesetting. The majority of the firesetters (89%, n = 16) indicated they had ignitedtheir fire(s) during adolescence due to boredom, peer pressure, to express feelings, orfor excitement. Firesetters and non-firesetters were similar in terms of socio-demographicand historical variables (e.g. age, number of siblings, from single parent households).However, firesetters self-reported significantly more diagnoses of behavioural problemsor convictions for a vandalism-related offence(s).

Gannon and Barrowcliffe (2012) found that relative to non-firesetters, firesetters scoredsignificantly higher on factors relating to antisocial behaviour on the FSS. In addition, thefiresetters self-reported significantly higher levels of fire fascination, behavioural propensity,and arousal on the FPS. However, this research can be improved. For example, the partici-pants were predominantly female university students, and aspects such as identificationwith fire, and attitudes towards fire were not measured.

More recently, Barrowcliffe and Gannon (2015) included additional measures in a studyconducted with a more representative sample of the UK population. Barrowcliffe andGannon (2015) randomly selected 10% of households from a high firesetting prevalencecommunity within Kent, UK (n = 5568), and hand delivered letters explaining the fireset-ting research. Participants accessed an online questionnaire and answered demographicquestions, self-reported their own deliberate firesetting, and completed five question-naires; the FSS, and the FPS (Gannon & Barrowcliffe, 2012), the Balanced Inventory of Desir-able Responding (BIDR version 6; Paulhus, 1984, 1988), the Fire Identification Scale (Gannon,Ó Ciardha, & Barnoux, 2011), and the Fire Attitude Scale (FAS; Muckley, 1997). The question-naire was completed by133 participants but 157 participants answered the firesettingquestion (2.8% partial, and 2.4% complete response rate, respectively; male n = 78,49.7%; female n = 79, 50.3%), and 18 participants (11.5%; male n = 11, 61.1%; female n =7, 38.9%) were classified as deliberate firesetters. Relative to the non-firesetters,

4 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

significantly more firesetters self-disclosed a family history of firesetting, reported a historyof self-harm, and had a father with a psychiatric illness. Firesetters also scored significantlyhigher on the FSS, the FPS, the Identification with Fire Scale, and the FAS compared to thenon-firesetters.

The format of the current study is similar to that of Barrowcliffe and Gannon (2015). Par-ticipants completed an online questionnaire relating to firesetting, and as fascination withfire (Barrowcliffe & Gannon, 2015; Gannon & Barrowcliffe, 2012; Rautaheimo, 1989), anti-social behaviour (Barrowcliffe & Gannon, 2015; Dolan, McEwan, Doley, & Fritzon, 2011;Gannon & Barrowcliffe, 2012), and fire interest have been found to play a role in bothadult (Barnoux, Gannon, & Ó Ciardha, 2015; Barrowcliffe & Gannon, 2015; Ó Ciardha &Gannon, 2012) and adolescent firesetting (MacKay et al., 2006; Watt, Geritz, Hasan,Harden, & Doley, 2015) measures associated with these traits were again included.However, research examining other psychological characteristics associated with appre-hended firesetters such as anger (Gannon et al., 2013; Rix, 1994), a lack of assertiveness,loneliness, social isolation (Hurley & Monahan, 1969; Inciardi, 1970; Jackson, Glass, &Hope, 1987; Noblett & Nelson, 2001; Rice & Chaplin, 1979), and boredom (Perrin-Wallqvist,Archer, & Norlander, 2004; Sapp, Huff, Gary, & Icove, 1999) have not been assessed in un-apprehended firesetting populations. Therefore demographic information in combinationwith additional scales examining the aforementioned psychological characteristics will beexamined with the aim of discriminating un-apprehended deliberate firesetters and non-firesetters.

Method

Participants

Participants were recruited through social media, and snow balling techniques. Twohundred and thirty-two people accessed an online questionnaire relating to firesetting.Of these 204 completed the questionnaire in full resulting in an 87.93% completionrate. Two hundred and twenty-five people (37 males, 188 females) answered the questionrelating to deliberate firesetting with an average age of 25.12 years (range 18–69 years).The majority of these participants indicated they were White (n = 175, 77.78%), of these75.43% (n = 132) identified themselves as White British, and 24.57% (n = 43) Whiteother. The majority of the participants were educated, only 3.11% (n = 7) indicated theyheld no qualifications, and 6.22% (n = 14) held only General Certificate of Secondary Edu-cations. The majority of participants had gained A level qualifications (or foreign equival-ent; n = 172; 76.44%) or a degree or higher (n = 30; 13.33%). Participants email addresseswere entered into a prize draw to win Amazon vouchers.

The measures

The online questionnaire contained three main sections a demographic and backgroundsection, firesetting disclosure, and numerous scales examining attitudes to fires, and person-ality variables: the demographic and background section contained questions relating togender, number of siblings, family background, psychiatric history, education level, andfamily background (e.g. parental psychiatric history, witnessing domestic violence,

PSYCHOLOGY, CRIME & LAW 5

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

family finances, and family history of firesetting). In the firesetting disclosure section partici-pants indicated whether they had ever ignited a fire to annoy other people, to relieveboredom, to create excitement, for insurance purposes, as a result of peer pressure, or toget rid of evidence. Fires set before the age of 10, ignited accidentally, or set as part ofan organized event (i.e. a bonfire) were to be excluded. Participants who indicated theyhad ignited a deliberate fire answered additional questions (e.g. forced choice questions)examining number of deliberate fires ignited, age at first and most recent firesetting inci-dent, formal apprehension or therapy relating to their firesetting, factors precipitating thefiresetting (i.e. intoxication, planning), modus operandi (i.e. the use of accelerants, ignitionpoints, distance of the fire from home), motivations, and targets of the deliberate fireset-ting, and response to the firesetting (i.e. attempts to extinguish the fire).

The final section of the questionnaire included various scales assessing fire interest andbehaviour (e.g. antisocial behaviour, boredom proneness, assertiveness, and anger) whichare presented in detail below. The scales were presented in a randomized order. Theinternal reliability αs are reported in accordance with George and Mallery’s (2003) criteria:≥.90 excellent, ≥.80 good, ≥.70 acceptable, and ≥.60 questionable.

The fire-related scales

There were five fire-related scales: the FSS and the FPS (Gannon & Barrowcliffe, 2012), theIdentification with Fire Questionnaire (Gannon et al., 2011), the FAS (Muckley, 1997), and theFire Interest Rating Scale (Murphy & Clare, 1996).

The Fire Setting ScaleThe 20-item FSS developed by Gannon and Barrowcliffe (2012) comprises two subscaleseach containing 10 items measuring AntiSocial Behaviour (e.g. ‘I am a rule breaker’) andFire Interest (e.g. ‘I get excited thinking about fire’). The items are rated using a 7-pointLikert scale (1 = not at all like me, 7 = very strongly like me). In a recent study, Barrowcliffeand Gannon (2015) noted that the internal consistency ranged from acceptable to excel-lent (overall α = .90, AntiSocial Behaviour α = .72, Fire Interest α = .92). See Appendix 1 forthe full version of the FSS. In the current study the internal consistency was also high(overall α = .91, AntiSocial Behaviour α = .85, Fire Interest α = .92).

The Fire Proclivity ScaleThe FPS also designed by Gannon and Barrowcliffe (2012) provides an indication of anindividual’s propensity to engage in deliberate firesetting. Participants are requested toimagine themselves as the firesetting protagonist in six hypothetical firesetting vignetteswhich vary in severity. Using a 5-point Likert scale participants responded to four ques-tions measuring: (1) fascination with the fire described in the scenario (1 = not at all fasci-nated to 5 = very strongly fascinated), (2) behavioural propensity to act similarly (1 =woulddefinitely not have done the same to 5 =would definitely have done the same), (3) generalarousal to the fire described in the scenario (1 =would not enjoy [watching it] at all to 5=would greatly enjoy [watching] it), and (4) general antisocialism (1 =would not enjoy[watching others’ reaction] at all to 5 = would greatly enjoy [watching others’ reaction]).The internal consistency of the FPS has previously been found to be acceptable (overallα = .93, fire fascination α = .86, behavioural propensity α = .66, fire arousal α = .81, and

6 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

general antisocialism α = .76; Barrowcliffe & Gannon, 2015). The internal consistency of theFPS was similar in the current study (overall α = .82, fire fascination α = .71, behaviouralpropensity α = .81, fire arousal α = .81, and general antisocialism α = .93). See Appendix2 for the full version of the FPS.

The Identification with Fire ScaleThe Identification with Fire Scale is a 10-item scale used to measure level of identificationwith fire (Gannon et al., 2011). Items such as ‘fire is almost part of my personality’ are ratedusing a 5-point Likert scale (1 = strong disagreement, 5 = strong agreement). Acceptableinternal consistency (α = .71) was found in a previous study by Barrowcliffe and Gannon(2015), but was questionable in the current study (α = .66).

The Fire Attitude ScaleThe FAS (Muckley, 1997) measures attitudes and beliefs about firesetting. Items such as‘the best thing about fire is watching it spread’ are rated using a 5-point Likert scale(1 = strong disagreement, 5 = strong agreement). Barrowcliffe and Gannon (2015) have pre-viously reported questionable internal consistency (α = .64). Slightly higher internal con-sistency was noted in the current study (α = .71).

The Fire Interest Rating ScaleThe Fire Interest Rating Scale (Murphy & Clare, 1996) contains 14 statements (e.g. ‘striking amatch to set fire to a building’) rated on a 7-point Likert scale (1 = extremely upsetting orfrightening, 4 = OK and it does not bother you, 7 = exciting, fun, or lovely). α informationfrom previous research is not available for The Fire Interest Rating Scale but the internalconsistency α was noted to be .82 in the current study.

Personality-related scales

The Novaco Anger Scale and Provocation InventoryThe Novaco Anger Scale and Provocation Inventory (NAS-PI; Novaco, 2003) contains 60items assessing four separate aspects associated with how anger is experienced; angercognitions (COG) ‘once something makes me angry, I keep thinking about it’,3 arousal(ARO) ‘When I get angry I stay angry for hours’, behavioural elements of anger (BEH)‘my temper is quick and hot’, and anger regulation (REG) ‘If I feel myself getting angry, Ican calm myself down’. Items are rated on a 3-point Likert scale (never, sometimes,always). The PI aspect of the NAS-PI contains 25 items associated with an individual’sability to tolerate provocation. Items (e.g. ‘someone else gets credit for work that youdid’, and ‘people who think they are better than you’) are responded to using a 4-pointLikert scale (1 = not at all angry, 4 = very angry).

The total scale and subscales have previously been found to have acceptable to excel-lent internal consistency when tested with a community sample (overall α = .92, COGα = .78, ARO α = .82, BEH α = .82, PI α = .92; Jones, Thomas-Peter, & Trout, 1999), and REGappears to exhibit acceptable reliability (α = .74) with a standardized sample (Novaco,2003). In the current study the NAS scale demonstrated slightly higher internal consist-ency, overall α = .92, COG α = .81, ARO α = .87, BEH α = .90, REG α = .95, and PI α = .95.

PSYCHOLOGY, CRIME & LAW 7

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

The Revised University of California Los Angeles (UCLA) Loneliness ScaleThe UCLA (Russell, Peplau, & Cutrona, 1980) is a 20-item self-report measure designed tomeasure social competence. Items such as ‘I lack companionship’ are responded to using a4-point Likert scale (1 = never, 4 = often). A reliability generalization study reported themean internal reliability coefficient α to be .87 (Vassar & Crosby, 2008). Similar psycho-metric properties were noted in the current study, α = .93.

The Simple Rathus Assertiveness Schedule – Short FormThe Simple Rathus Assertiveness Schedule – Short Form (Jenerette & Dixon, 2010)consists of 19 items (e.g. ‘I am quick to say what I think’) rated on a 6-point Likert scale(1 = very much unlike me, 6 = very much like me). Jenerette and Dixon (2010) reportedthat their scale had good reliability (α = .81), this was also evident in the current study(α = .82).

Nowicki Strickland Locus of ControlThe Nowicki Strickland Locus of Control (Nowicki, 1976) scale measures how much a par-ticipant feels they are in control of the events around them. The 40 items in the scale (e.g.‘Are some people just born lucky’) are responded to with either a yes or no answer. Thescale has been noted to have levels of internal consistency ranging between α = .66and α = .75 (Duke & Nowicki, 1973), and similarly the internal consistency in the currentstudy (α = .69) falls within this range.

Boredom Proneness Scale – Short FormThe Boredom Proneness Scale – Short Form (Vodanovich, Wallace, & Kass, 2005) contains12 items measuring internal (e.g. ‘I find it easy to entertain myself’) and external (e.g. ‘Itseems that the same things are on television or the movies all the time; it’s getting old’)factors relating to boredom. The items are measured using a 7-point Likert scale (1 =strongly disagree, 7 = strongly agree). The short form has been noted to have acceptablereliability (α = .70; Hopley & Nicki, 2010), and α = .73 in the current study.

Measure of Criminal Attitudes and Associates – Part BThe Measure of Criminal Attitudes and Associates (M-CAA-Part B; Mills & Kroner, 1999) is a46-item scale (agree/disagree) which measures attitudes towards violence (e.g. ‘Sometimesyou have to fight to keep your self-respect’), entitlement (e.g. ‘It is wrong for a lack ofmoney to stop you from getting things’), antisocial intent (e.g. ‘Rules will not stop mefrom doing what I want’), and associates (e.g. ‘I have committed a crime with friends’).The psychometric properties of the M-CAA-Part B have previously been reported torange between acceptable and good (Part B total α = .75, violence α = .80, antisocialindex α = .72, and associates α = .82) with the exception of entitlement (α = .63; Mills,Kroner, & Forth, 2002). In the current study the reliability α was .81 for the completeM-CAA-Part B (violence α = .76, entitlement α = .72, antisocial index α = .41, and antisocialassociates α = .01). The negligible α for the antisocial associates may be as a result of only asmall minority of participants (n = 5, 2.2%) having ever been arrested.

8 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

Impression Management

Balanced Inventory of Desirable RespondingThe BIDR (Paulhus, 1984, 1988) is in total a 40-item scale. The Impression Management (IM)section makes up half of the scale where items (e.g. ‘I never swear’) are rated on a 5-pointLikert scale (1 = not true, 5 = very true). The BIDR IM scale has good internal consistency (αranging from .75 to .86; Paulhus, 1988). In a community sample internal consistency wasnoted to be α = .83 (Barrowcliffe & Gannon, 2015). However in the current communitysample the α was noted to be lower (α = .66).

Procedure

The research was ethically approved by the University’s Research Ethics Committee (Ref20142842). Participants completed the online questionnaire in their own time. To encou-rage participation, participants’ email addresses were entered into a prize draw to winAmazon vouchers. Participants viewed an information sheet before the start of thestudy, and were informed that continuing with the study indicated consent. Participantswere requested not to disclose any personally identifying information about themselvesor any fires they may have started. In order to ensure anonymity IP addresses were notrecorded. At the end of the questionnaire participants were thanked, and a writtendebrief appeared explaining the purpose of the research, and reiterating that the infor-mation provided would remain anonymous.

Results

Firesetting prevalence and features

Forty participants (17.78%) indicated that they had ignited a deliberate fire but had notbeen formally apprehended for their actions. The majority of these firesetters reportedthat they were White British (n = 26, 65%), and all held a formal qualification (e.g. Alevels or higher). Overall the majority of the firesetters were female (n = 25, 62.5%). Ofthe 37 males and 188 females who participated 40.54% (n = 15) of males and 13.31%(n = 25) of females were classified as firesetters. Key demographics can be found inTable 1.

Firesetters self-reported igniting their most recent fire between the ages of 10 and 37years (median age 16 years). The majority (85%, n = 34) of firesetters reported igniting theirmost recent fire between 10 and 18 years of age, and only 15% (n = 6) igniting fires duringadulthood (range 20–37 years). Only one firesetter (2.5%) ignited their first fire inadulthood.

Fifteen firesetters ignited only one deliberate fire (37.5%). However the majority of fire-setters ignited multiple fires (see Table 2). None of the firesetters reported holding anyconvictions for arson but three firesetters reported holding convictions for either aviolent crime, antisocial behaviour, or theft. None of the firesetters reported havingreceived therapy for their firesetting behaviour. The majority of firesetters (n = 28, 70%)reported igniting a fire within one mile of their home (e.g. walking distance). One firesetterreported being under the influence of drugs, and five firesetters (12.5%) claimed to be

PSYCHOLOGY, CRIME & LAW 9

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

Table 1. Firesetters and non-firesetters historical characteristics and demographics.

Variable

Firesetters Non-firesetters(N = 40) (N = 185)M (SD) M (SD)

DemographicsAge 24.00 (9.00) 25.50 (12.73)Siblings (number) 2.50 (1.20) 2.62 (1.12)

Percentage yes (n) Percentage yes (n)

Males 37.5 (15) 11.9 (22)Females 62.5 (25) 88.1 (163)White British 65.0 (26) 57.3 (106)White Other 20.0 (8) 18.9 (35)Formal qualifications 100 (40) 96.2 (178)History of enuresis 7.5 (3) 4.3 (8)Psychiatric illness diagnosis 32.5 (13)** 14.1 (26)Physical disability diagnosis 5.0 (2) 1.6 (3)Behavioural problem diagnosis 12.5 (5)** 0 (0)Suspension from school 32.5 (13)** 4.3 (8)Expulsion from school 10.0 (4) 3.2 (6)History of suicide 17.5 (7)* 6.5 (12)History of self-harm 35.0 (14) 21.6 (40)Criminal convictions 7.5 (3) 1.1 (2)Experimented with fire before the age of 10 57.5 (23)** 24.3 (45)Family backgroundLack of money (i.e. sometimes not enough money for food) 25.0 (10) 14.6 (27)Witnessed domestic violence 25.0 (10) 25.9 (48)Mother diagnosed with a psychiatric illness 30 (12) 17.8 (33)Father diagnosed with a psychiatric illness 12.5 (5) 10.3 (19)A family member also ignited a deliberate fire 15 (6)** 3.2 (6)

Note: χ2 with 95% confidence.*p≤ .05.**p≤ .01.

Table 2. Deliberate firesetting offence characteristics.Offence characteristics Percentage yes (n)

Number of deliberate fires ignitedOne 37.5 (15)Two 27.5 (11)Three 17.5 (7)Four or more 17.5 (7)Ignition point and targetOne ignition point 67.5 (27)Multiple ignition points 32.5 (13)Ignited countryside (e.g. grass/shrubbery) 27.5 (11)Paper, books, or newspapers 25.0 (10)Ignited a bin outside 22.5 (9)Ignited a bin inside 12.5 (5)Ignited clothing 12.5 (5)Ignited a toilet roll dispenser 12.5 (5)General rubbish 7.5 (3)Furniture 5.0 (2)Ignited an unoccupied car 2.5 (1)Ignited an animal which was alive 2.5 (1)Ignited a house knowing it was occupied 2.5 (1)Fires ignited alone or with accomplicesIgnited fire alone 27.5 (11)Ignited fire with 1 other person 12.5 (5)Ignited fire with 2 other people 20.0 (8)Ignited fire with 3+ people 40.0 (16)

Note: Ignition targets do not add up to 100% due to multiple targets.

10 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

under the influence of alcohol during ignition. The majority of firesetters ignited their fireswith other people (n = 29, 72.5%). Table 2 contains further offence characteristics.

Participants were requested to list the motivations behind their firesetting (see Table 3).The majority of firesetters (n = 27, 67.5%) reported multiple motivations. The predominantmotivations behind firesetting were experimentation and curiosity (n = 26, 65%), and toalleviate boredom or create excitement (n = 27, 67.5%). None of the firesetters were motiv-ated by revenge, and nine firesetters (22.5%) stated they experienced a love of fire.

Just under a third of firesetters ignited grass, shrubbery or dry leaves (n = 11, 27.5%) and10 firesetters (25%) ignited paper, books, or newspapers. Igniting waste paper baskets andbins inside (n = 5, 12.5%) and outside (n = 9, 22.5%) were also common targets (see Table2). Although the majority of firesetters (n = 30, 75%) took part in extinguishing the fire, fourfiresetters (10%) indicated that the Fire Service extinguished their fires. In terms of preven-tative measure, 35% of firesetters (n = 14) indicated that having better fire safety knowl-edge (e.g. being aware of the dangers, and increased knowledge of how fire develops)would have prevented them from firesetting but 45% (n = 18) indicated that nothingwould have prevented their firesetting. Of the participants who indicated that havingbetter fire safety knowledge would have prevented them from firesetting seven ignitedjust one fire but the remaining seven firesetters ignited multiple fires (two fires [n = 3],three fires [n = 2], or four or more fires [n = 2]).

Firesetter and non-firesetter characteristic comparisons

Demographic and historical variablesUnivariate comparisons (see Table 1) revealed that firesetters and non-firesetters could notbe significantly differentiated on the majority of demographic, or historical variables (e.g.age, number of siblings, history of enuresis, formal qualifications, physical disability, historyof self-harm, criminal convictions, witnessing domestic violence, or parental psychiatrichistory).

Relative to non-firesetters, firesetters were more likely to report a diagnosis of a psychia-tric illness,4 χ2 (1, n = 225) = 6.58, p≤ .01, φ = .19, and a diagnosis of a behavioural disorder(e.g. Attention Deficit Hyperactivity Disorder (ADHD)) χ2 (1, n = 225) = 18.25, p < .01, φ= .32. All of the firesetters with a behavioural disorder began firesetting in childhood

Table 3. The motivations behind deliberate firesetting.Motivation Percentage yes (n)

Curiosity or experimenting with fire 65.0 (26)To create fun/excitement or alleviate boredom 67.5 (27)Love fire 22.5 (9)Dared or pranked 20.0 (8)Vandalism 10.0 (4)Going along with friends 5.0 (2)Stressed or frustrated 5.0 (2)Problems at home or school 2.5 (1)Protecting themselves 2.5 (1)Revenge 0Insurance payout or financial gain 0Covering up another crime/destroying evidence 0

Note: Motivations do not add up to the number of firesetters as many firesetters (n = 27)indicated multiple motives.

PSYCHOLOGY, CRIME & LAW 11

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

and adolescence (10–15 years of age) and ignited more than one fire (two fires n = 3, threefires n = 1, five or more fires n = 1). Firesetters were also more likely to have experimentedwith fire before the age of 10 years χ2 (1, n = 225) = 15.63, p < .01, φ = .28, and to have beensuspended from school χ2 (1, n = 225) = 27.61, p < .01, φ = .37. In addition relative to non-firesetters, firesetters reported having engaged in suicide attempts χ2 (1, n = 225) = 3.83,p≤ .05, φ = .15, and having a family member who had also ignited a deliberate fire χ2

(1, n = 201) = 9.60, p < .01, φ = .25.

Questionnaire measuresThe IM subscale of the BIDR significantly correlated with the FPS, the FAS, the NAS-PI, theRevised UCLA Loneliness Scale, and the M-CAA. However, for all of these scales the fireset-ters’ scale scores did not significantly correlate with the IM subscale when computed sep-arately for firesetters and non-firesetters.

Mean scale scores were calculated separately for the firesetters, and non-firesetters (seeTable 4). Separate one-way between-groups multivariate analysis of variance (MANOVA)were conducted to establish any differences between firesetters and non-firesetters onthe FSS, the FPS, the NAS, and the M-CAA. After checking assumptions for normality, lin-earity, outliers, multicollinearity, and homogeneity of variance–covariance the separateMANOVA analyses confirmed that firesetters scored significantly higher compared to

Table 4. Reliability and scale scores for self-reported deliberate firesetters and non-firesetters.

Scale Cronbach α

Firesetters Non-firesetters ScalerangeM SD M SD

FSS .91 68.40*** (21.87) 50.70 (17.38) 20–140Behavioural items .85 33.10*** (11.43) 24.59 (8.89) 10–70Fire Interest items .92 35.30*** (13.76) 26.11 (12.25) 10–70

FPS .82 56.43*** (18.49) 44.81 (13.03) 24–120Fire Fascination .71 16.55*** (6.16) 13.13 (4.56) 6–30Behavioural Propensity .81 13.85*** (4.87) 10.17 (3.15) 6–30Fire Arousal .81 16.00*** (5.73) 12.63 (4.25) 6–30Antisociality .93 10.03* (3.49) 8.88 (3.16) 6–30

Identification with Fire Scale .66 19.74 (5.73) 18.08 (4.23) 10–50FAS .71 53.28 (11.46) 52.00 (7.11) 20–100The Fire Interest Rating Scale .82 45.97** (10.53) 40.59 (7.84) 14–98NAS-PI T scores .92 115.49*** (13.17) 105.84 (14.36)Cognition (COG) .81 31.95*** (4.58) 28.67 (4.62) 20–80Arousal (ARO) .87 30.38** (5.42) 27.52 (5.64) 21–80Behavioural (BEH) .90 28.13*** (6.25) 24.18 (5.80) 29–80Regulation (REG) .95 25.03 (3.87) 25.46 (3.52) 20–80PI .95 68.37* (12.72) 62.71 (14.59) 20–79

The Revised UCLA Loneliness Scale .93 42.06 (13.39) 37.51 (9.75) 20–80The Simple Rathus Assertiveness Schedule – ShortForm

.82 69.51 (14.06) 66.35 (12.22) 19–114

Nowicki Strickland Locus of Control .69 14.74 (5.99) 13.34 (5.77) 0–40Boredom Proneness Scale – Short Form .73 44.42* (10.02) 40.59 (7.51) 12–84M-CAA .81 19.95*** (7.21) 15.07 (5.41) 0–46Violence .76 4.72** (3.32) 2.93 (2.20) 0–12Entitlement .72 5.72** (3.16) 4.43 (2.61) 0–12Antisocial .41 5.51*** (1.73) 4.18 (1.71) 0–12Associates .01 4.00* (1.50) 3.53 (1.23) 0–10

BIDR – IM .66 54.21 (7.11) 59.85 (9.37) 20–100

*p < .05.**p < .01.***p < .001.

12 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

non-firesetters on the combined firesetting scale F(2,216) = 16.27, p < .01; Wilks’ Λ = .87;h2p = .13. When the results of the dependent variables were considered separately both

of the subscales were also significant; Behavioural subscale F(1,217) = 26.79, p < .01;h2p = .11, Fire Interest subscale F(1,217) = 17.56, p < .01; h2

p = .08. Similarly, the firesettersalso scored significantly higher compared to the non-firesetters on the combined FPS F(4,215) = 10.24, p < .01; Wilks’ Λ = .84; h2

p = .16 and all of the subscales (Fascination F(1,218) = 16.06, p < .01; h2

p = .07, Behavioural Propensity F(1,218) = 35.78, p < .01;h2p = .14, Arousal Index F(1,218) = 17.98, p < .01; h2

p = .08, and Antisocial Index F(1,218)= 4.10, p < .05; h2

p = .02).A separate MANOVA also showed that firesetters scored significantly higher on the

combined NAS5 F(5,195) = 3.53, p < .01; Wilks’ Λ = .92; h2p = .08, and the majority of its sub-

scales (COG F(1,199) = 15.26, p < .01; h2p = .07, ARO F(1,199) = 9.33, p < .01; h2

p = .05, BEH F(1,199) = 12.72, p < .01; h2

p = .06, and PI F(1,199) = 5.61, p < .05; h2p = .03). The subscale

relating to the regulation of anger (REG) was not significant.The MANOVA for the M-CAA-Part B revealed that relative to the non-firesetters the fire-

setters scored significantly higher on the combined subscales of the M-CAA-Part B, F(4,211) = 6.81, p < .01; Wilks’ Λ = .89; h2

p = .11, and also scored significantly higher on itssubscales (Violence F(1,214) = 17.13, p < .01; h2

p = .07, Entitlement F(1,214) = 7.18, p< .01; h2

p = .03, Antisocial Index F(1,214) = 19.26, p < .01; h2p = .08, and Associates F

(1,214) = 4.37, p < .05; h2p = .02).

Independent samples t-tests confirmed that firesetters scored significantly higher com-pared to non-firesetters on the Fire Interest Rating Scale t(47.71) = 3.02, p < .01, d = .87(two-tailed, mean difference = 5.39, 95% CI: 1.79, 8.98). Firesetters also scored significantlyhigher on the Boredom Proneness Scale t(46.96) = 2.22, p < .05, d = .65 (two-tailed, meandifference = 3.83, 95% CI: 0.36, 7.30). There were no significant differences between thescores of firesetters and non-firesetters on the Identification with Fire Scale, the FAS, theRevised UCLA Loneliness Rating Scale, the Simple Rathus Assertiveness Scale, or theNowicki Strickland Locus of Control Scale.

Classifying firesetters and non-firesettersA total of 12 variables differentiated the deliberate firesetters and the non-firesetters. Dueto small sample sizes two separate logistic regressions were conducted, one to assess thestatic variables and one relating to the dynamic variables. There were six static variables; adiagnoses of a psychiatric illness, a diagnosis of a behavioural problem, suspension fromschool, history of suicide attempts, experimentation with fire before the age of 10 years,and having a family member who had also ignited a deliberate fire. However due to asmall number of participants (n = 5) reporting a diagnosis of a behavioural problem thisvariable was omitted from the analysis. The complete model was significant χ2 (5, n =201) = 41.81, p < .01, and therefore able to distinguish between the self-reported fireset-ters and non-firesetters. As a whole the model explained between 18.8% (Cox and SnellR2) and 33.0% (Nagelkerke R2) of the variance in firesetting status, and correctly classified87.1% of cases overall.

Three independent variables, having been suspended from school, experimenting withfire before the age of 10 years, and having a family history of firesetting were statisticallysignificant contributors to the model with odds ratios of 0.10, 0.32 and 0.23, respectively.Thus participants who had been suspended from school, experimented with fire before

PSYCHOLOGY, CRIME & LAW 13

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

age 10, or had a family member with a history of deliberate firesetting were more likely tobe classified as deliberate firesetters (see Table 5).

In terms of the dynamic variables, firesetters and non-firesetters scored significantly dif-ferently on six scale measures: the FSS, the FPS, the Fire Interest Rating Scale, the NAS-PI,the Boredom Proneness Scale, and the M-CAA. The complete model was significant χ2 (6,n = 200) = 33.10, p < .01, and therefore able to distinguish between the self-reported fire-setters and non-firesetters. As a whole the model explained between 15.3% (Cox and SnellR2), and 24.5% (Nagelkerke R2) of the variance in firesetting status, and correctly classified83.5% of cases. However, none of the variables were individually statistically significantcontributors to the model overall (see Table 6).

Discussion

The current study extends the firesetting literature by examining the behaviour and per-sonality characteristics of un-apprehended firesetters. A total of 17.78% of the participantswere classified as deliberate firesetters. This prevalence rate is considerably higher thanthe prevalence rate of 1–1.13% in the USA study (Blanco et al., 2010; Vaughn et al.,2010), and 11% (Gannon & Barrowcliffe, 2012), and 11.5% prevalence rates in UK commu-nity studies (Barrowcliffe & Gannon, 2015). One explanation for the increase in prevalencerate is as a result of the way in which participants were recruited. For example participantsdid not meet the researchers face to face but were instead recruited online and assuredthat their responses could not be traced back to them.

Igniting fires close to home is a feature associated with apprehended firesetters (Brad-ford, 1982; Rautaheimo, 1989; Wachi et al., 2007), and community firesetters in both thisstudy, and that of Barrowcliffe and Gannon (2015). Apprehended firesetting populations

Table 5. Logistic regression predicting firesetter status based on static variables.

β SE Wald df p Odds ratio

95% CI for oddsratio

Lower Upper

Diagnosis of a psychiatric illness −0.58 0.59 0.96 1 .33 0.56 0.18 1.78Suspension from school −2.35 0.58 16.40 1 <.001 0.10 0.03 0.30History of suicide attempts −0.66 0.74 0.81 1 .37 0.52 0.12 2.18Experimented with fire before age 10 −1.15 0.47 5.93 1 .02 0.32 0.13 0.80Family history of deliberate firesetting −1.46 0.73 3.98 1 .05 0.23 0.06 0.98Constant 3.24 1.03 10.05 1 <.001 25.8

Table 6. Logistic regression predicting firesetter status based on dynamic variables.

β SE Wald df p Odds ratio

95% CI for oddsratio

Lower Upper

FSS 0.03 0.02 2.96 1 .09 1.03 1.00 1.06FPS 0.01 0.02 0.14 1 .71 1.01 0.97 1.05Fire Interest Rating Scale 0.01 0.03 0.10 1 .76 1.01 0.96 1.06NAS-PI (T scores) 0.01 0.01 2.45 1 .19 1.01 1.00 1.03Boredom Proneness Scale 0.01 0.03 0.06 1 .81 1.01 0.95 1.06M-CAA 0.05 0.04 1.77 1 .18 1.05 0.98 1.14Constant −8.16 2.06 15.70 1 .00 0.00

14 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

and the un-apprehended firesetters in this current study also share some similar charac-teristics. For example apprehended firesetters tend to have a history of self-harm andsuicide (Jayaraman & Frazer, 2006; McKerracher & Dacre, 1966; Noblett & Nelson, 2001;O’Sullivan & Kelleher, 1987; Swaffer & Hollin, 1995). Similarly the community firesetterswere also noted to have significantly more suicide attempts compared to the non-fireset-ters. In addition relative to non-firesetters, community firesetters reported significantlymore diagnoses of psychiatric illness. However it is unclear when the diagnosis tookplace (e.g. before or after the firesetting) or if the firesetters were having symptoms atthe time of ignition.

Firesetting offenders relative to other non-firesetting offenders have been found to bedistinguishable based on fire-related factors such as fire interest, increased anger cogni-tions, and susceptibility to anger provocation (Gannon et al., 2013). Fire interest hasbeen found to increase the likelihood of firesetting in both adults (Barnoux et al., 2015;Ó Ciardha & Gannon, 2012), and adolescents (MacKay et al., 2006; Watt et al., 2015). Simi-larly fascination with fire (Rautaheimo, 1989) and antisocial behaviour (Dolan et al., 2011)are also linked to firesetting behaviour. Although the current study with un-apprehendedfiresetters is not nationally representative, and is relatively small scale it offers an insightinto the relevance of fire interest and fascination, anger cognitions, and antisocial behav-iour as these factors significantly differentiated un-apprehended firesetters and non-fire-setters. Un-apprehended firesetters in the current study held more positive attitudestowards fires such as interest around fire which is consistent with the literature relatingto the implicit theories associated with adult apprehended firesetters (O’Ciardha &Gannon, 2012) and adolescent firesetters (Watt et al., 2015).

Apprehended firesetters are noted to be unskilled, and have low IQ (Bradford, 1982;Harmon et al., 1985; Lewis & Yarnell, 1951; Rautaheimo, 1989). In contrast with the appre-hended firesetting literature, but in line with the research associated with un-appre-hended firesetters (Barrowcliffe & Gannon, 2015) all of the un-apprehended firesettersin the current study were educated, and held at least A-level (Advanced level) UK qualifi-cations. However, it is acknowledged that the participant sample were generally educatedwith only seven participants (3.11%) holding no qualifications.

The majority of the firesetters (85%, n = 34) in this community study reported ignitingfires between the ages of 10–18 years. Similarly the majority of firesetters in the NESARCstudy (Blanco et al., 2010; Vaughn et al., 2010) and both the studies with UK un-appre-hended firesetters (Barrowcliffe & Gannon, 2015; Gannon & Barrowcliffe, 2012) ignitedfires during adolescence. However in contrast to the literature associated with appre-hended adolescent and adult firesetters (Bradford, 1982; Muller, 2008; Pettiway, 1987;Räsänen et al., 1995; Rautaheimo, 1989) the majority of the firesetters in the currentstudy were female (n = 25, 62.5%). Yet this is not surprising as significantly morefemales (n = 188, 83.56%) participated compared to males (n = 37, 16.44%). It is worthnoting that 40.54% (n = 15) of the male participants and 13.31% (n = 25) of female partici-pants indicated that they had ignited a fire which matched the criteria for deliberatefiresetting.

Both male and female offenders in general (Andrews & Bonta, 2010; Andrews et al.,2012) and apprehended firesetters are noted to have substance abuse issues (Jayaraman& Frazer, 2006), and/or issues with alcohol (Bourget & Bradford, 1989; Rautaheimo, 1989).However none of Barrowcliffe and Gannon’s (2015) un-apprehended firesetters cited that

PSYCHOLOGY, CRIME & LAW 15

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

alcohol or drugs played a role in their firesetting. Similarly in the current community studyjust one firesetter (2.5%) self-reported being under the influence of substances at the timeof ignition, and five firesetters (12.5%) indicated they were slightly to moderately intoxi-cated at the time of ignition. As individuals under the influence are unlikely to becompos mentis enough to cover their tracks and evade detection, it is likely thatalcohol and drug issues are over represented in the apprehended firesetting population.

The apprehended firesetting literature cites revenge as the predominant motivationbehind firesetting (Lewis & Yarnell, 1951; O’Sullivan & Kelleher, 1987; Swaffer & Hollin,1995). However revenge was not cited as a motivation in this study, or previous researchwith un-apprehended firesetters (Barrowcliffe & Gannon, 2015; Gannon & Barrowcliffe,2012). Revenge fires are likely to target an individual or their property and it is thereforefeasible to assume that they are larger more destructive fires which have an increased like-lihood of coming to the attention of the authorities, and leading to apprehension. Insteadcuriosity, excitement and alleviating boredom were the most common motivations for theun-apprehended firesetters in the current study and previous un-apprehended firesettingresearch (Barrowcliffe & Gannon, 2015; Gannon & Barrowcliffe, 2012). Interestingly, in thecurrent study, boredom was a motivation associated only with adolescent firesetters. Forthis reason encouraging adolescents to attend youth engagement programmes (e.g. afterschool activities and youth clubs) may help to alleviate boredom and prevent these indi-viduals from deliberately igniting fires.

It is concerning that a reasonable percentage of the firesetters ignited fires inside build-ings (n = 5, 12.5%). Similarly concerning is that although the majority of firesetters (n = 30,75%) extinguished their fires, four firesetters (10%) indicated that the Fire Service inter-vened. Presumably the fires extinguished by the Fire Service were larger, and moredestructive in nature. Worryingly 45% (n = 18) of firesetters indicated that nothingwould have prevented them from deliberately igniting a fire. However, some comfortcan be found in the fact that 35% of firesetters (n = 14), indicated that having better firesafety knowledge, such as being aware of the dangers of fire, and increased knowledgeof how a fire develops would have prevented them from firesetting.

ADHD has been cited as a factor in firesetting, but the data are limited (see Dolan et al.,2011). For example, McCardle, Lambie, and Barker-Collo (2004) found that just over half(53%) of their adolescent male firesetters in New Zealand had a diagnosis of ADHD.However this information was obtained from parent/caregivers rather than relying onmedical records. In contrast none of the un-apprehended firesetters in the Barrowcliffeand Gannon (2015) study self-reported a behavioural disorder diagnosis. In the currentcommunity research five (12.5%) of the firesetters self-reported a behavioural disorderand firesetting was predominantly associated with younger firesetters. For example themajority of firesetters who reported a diagnosis of a behavioural disorder ignited theirfirst and last fires in childhood or adolescence (10–18 years of age), with just one firesetterigniting a fire at 20 years old. Interestingly firesetters diagnosed with a behavioural dis-order ignited multiple fires which supports the idea that firesetting may be an advancedlevel of antisocial behaviour (Forehand, Wierson, Frame, Kempton, & Armistead, 1991) thatwarrants further research.

When comparing un-apprehended firesetters and non-firesetters previous communityresearch has found that relative to non-firesetters, firesetters were significantly less likelyto have experimented with fire before the age of 10 (Barrowcliffe & Gannon, 2015).

16 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

Barrowcliffe and Gannon (2015) suggested that firesetters may hold restricted experiencesin manipulating fire as children which feeds into their motivation to misuse fire later on.However in contrast, in the current community sample firesetters were significantlymore likely to have experimented with fire before 10 years of age. It is thought that fireinterest is common in childhood but by the age of 10 the majority of children have areasonable understanding of fire safety (Dolan et al., 2011). However, early firesetting inchildhood is believed to be a significant predictor of subsequent fire involvement forboth patients and non-patients (Kolko, 2001). Dolan et al. (2011) suggests that firesettingdevelops into a problematic issue for children who lack adequate supervision. The partici-pants in the current study were not asked to comment on the supervision they received asa child but this may offer an explanation as to how their firesetting remained un-noticed.Clearly, it would be beneficial for future research to be conducted with larger samples totruly assess the effects of family background and childhood fire experiences on sub-sequent firesetting behaviour.

In line with previous community research (Barrowcliffe & Gannon, 2015) the currentcommunity firesetters were also more likely to have a family member who had igniteda deliberate fire. These findings further support theoretical models of firesettingsuggesting that there is a social learning aspect associated with firesetting behaviour(see Gannon et al., 2012; Jackson et al., 1987).

As this research was conducted via social media the recruitment rate cannot be deter-mined and therefore it is not possible to comment on any sample selection biases.However, it is acknowledged that the gender participation bias (high female to male par-ticipation rate) is a research limitation. Other researchers have also found that relative tomales, high female participation rates are a common research problem (Sax, Gilmartin, Lee,& Hagedorn, 2008; Underwood, Kim, & Matier, 2000). In addition the findings of the currentstudy are limited by self-report measures. It is also acknowledged that unintentionalmemory recollection failures may have occurred as the majority of un-apprehended fire-setters (85%) were retrospectively commenting on their firesetting behaviour in adoles-cence. Future research concerned with un-apprehended firesetting should perhaps aimto recruit younger participants.

The factors which significantly differentiated the firesetters, and non-firesetters wereentered into a logistic regression to gauge their ability to predict firesetting status. Twoseparate logistic regressions were conducted to assess the predictive ability of fivestatic variables (a diagnoses of a psychiatric illness, suspension from school, history ofsuicide attempts, experimentation with fire before the age of 10 years old, and having afamily member who has also ignited a deliberate fire), and six dynamic variables (theFSS, the FPS, the Fire Interest Rating Scale, the NAS-PI, the Boredom Proneness Scale,and the M-CAA). Only three static variables successfully predicted firesetting status (sus-pension from school, experimentation with fire before the age of 10 years old, andhaving a family member who has also ignited a deliberate fire). The current research war-rants further investigation, but supports the findings that previous firesetting incidencesare the best predictors of future firesetting in both child, adolescent, and adult firesetters(Edwards & Grace, 2014; Kennedy, Vale, Khan, & McAnaney, 2006; Kolko, 2001).

The literature associated with un-apprehended UK firesetters is limited. Relative to theirapprehended counterparts, un-apprehended firesetters appear to be highly educated, andpossess the ability to evade detection. Therefore it is inappropriate to apply all of the

PSYCHOLOGY, CRIME & LAW 17

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

research associated with apprehended firesetters to un-apprehended firesetters, instead itwould be beneficial to conduct further research in this area. In particular it is interesting tonote that some un-apprehended firesetters highlighted factors which they believe wouldhave prevented them from firesetting (e.g. better fire education). Incorporating such pro-grammes into educational curriculums is a step closer to reducing the incidences, injuries,and fatalities caused as a result of deliberate firesetting.

Notes

1. Number of male and female firesetters does not add up to total number of cases proceededagainst in court as in 10 adult cases and 3 juvenile cases the gender was unspecified.

2. In the UK the law assumes that children under the age of 10 are too young to understand themorality of their behaviour and therefore cannot receive a criminal conviction (Gov.UK, 2015).

3. Sample items from the NAS-PI copyright © 2003 by Western Psychological Services. Reprinted byE. Barrowcliffe, University of Kent, for scholarly display purposes by permission of the publisher,WPS, 625 Alaska Avenue, Torrance, California 90503, USA. Not to be reprinted in whole or in partfor any additional purpose without the expressed, written permission of the publisher (right-s@wpspublish.com). All rights reserved.

4. Psychiatric disorder included depression, schizophrenia, obsessive compulsive disorder, eatingdisorders and anxiety disorders.

5. Novaco t-score conversions were used in the analysis.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Andrews, D. A., & Bonta, J. (2010). The psychology of criminal conduct (5th ed.). New Providence, NJ:LexisNexis Matthew Bender.

Andrews, D. A., Guzzo, L., Raynor, P., Rowe, R., Rettinger, L. J., Brews, A., & Wormith, J. S. (2012). Are themajor risk/need factors predictive of both female and male reoffending? A test with the eightdomains of the Level of Service/Case Management Inventory. International Journal of OffenderTherapy and Comparative Criminology, 56, 113–133. doi:10.1177/0306624X10395716

Barnoux, M., Gannon, T. A., & Ó Ciardha, C. (2015). A descriptive model of the offence chain for impri-soned adult male firesetters (descriptive model of adult male fi(descript). Legal and CriminologicalPsychology, 20, 48–67. doi:10.1111/lcrp.12071

Barrowcliffe, E. R., & Gannon, T. A. (2015). The characteristics of un-apprehended firesetters living inthe UK community. Psychology, Crime and Law. doi:10.1080/1068316X.2015.1054385

Blanco, C., Alegria, A. A., Petry, N. M., Grant, J., Simpson, H. B., Liu, S.,… Hasin, D. (2010). Prevalenceand correlates of firesetting in the US: Results from the national epidemiologic survey on alcoholand related conditions (NESARC). Journal of Clinical Psychiatry, 71(9), 1218–1225. doi:10.4088/JCP.08m04812gry

Bourget, D., & Bradford, J. M. W. (1989). Female arsonists: A clinical study. Bulletin of the AmericanAcademy of Psychiatry and the Law, 17(3), 293–300.

Bradford, J. M. W. (1982). Arson: A clinical study. Canadian Journal of Psychiatry, 27, 188–193.Davis, J. A., & Lauber, K. M. (1999). Criminal behavioral assessment of arsonists, pyromaniacs, andmul-

tiple firesetters. Journal of Contemporary Criminal Justice, 15(3), 273–290. doi:10.1177/1043986299015003005

Dennet, M. F. (1980). Fire investigation. Oxford: Pergamon Press.

18 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

Department for Communities and Local Government. (2011). The economic cost of fire: Estimates for2008. London: Crown Copyright.

Department for Local Communities and Local Government. (2014). Fire statistics: Great Britain April2012 to March 2013. London: Crown Copyright.

Dickens, G. L., Sugarman, P. A., & Gannon, T. A. (Eds). (2012). Firesetting and mental health, theoryresearch and practice (pp. 3–27). London: RCPsych.

Dolan, M., McEwan, T. E., Doley, R., & Fritzon, K. (2011). Risk factors and risk assessment in Juvenile fire-setting. Psychiatry, Psychology and Law, 18(3), 378–394. doi:10.1080/13218719.2011.559154

Doley, R. (2003). Making sense of arson through classification. Psychiatry, Psychology and Law, 102,346–352. doi:10.1375/pplt.2003.10.2.346

Duke, M. P., & Nowicki, S. (1973). Personality correlates of the Nowicki-Strickland locus of control scalefor adults. Psychological Reports, 33, 267–270.

Edwards, M. J., & Grace, R. C. (2014). The development of an actuarial model for arson recidivism.Psychiatry, Psychology and Law, 21(2), 218–230. doi:10.1080/13218719.2013.803277

Fessler, D. T. (2006). A burning desire: Steps towards an evolutionary psychology of fire learning.Journal of Cognition & Culture, 6(3/4), 429–451. doi:10.1163/156853706778554986

Forehand, R., Wierson, M., Frame, C. L., Kempton, T., & Armistead, L. (1991). Juvenile firesetting: Aunique syndrome or an advanced level of antisocial behavior? Behaviour Research and Therapy,29(2), 125–128. doi:10.1016/0005-7967(91)90040-A

Gannon, T. A. (2010). Female arsonists: Key features, psychopathologies, and treatment needs.Psychiatry, 73(2), 173–189. doi:10.1521/psyc.2010.73.2.173.

Gannon, T. A., & Barrowcliffe, E. R. (2012). Firesetting in the general population: The development andvalidation of the fire setting and fire proclivity scales. Legal and Criminological Psychology, 17(1),105–122. doi:10.1348/135532510X523203

Gannon, T. A., Ó Ciardha, C., & Barnoux, M. (2011). The identification with fire questionnaire(Unpublished manuscript). CORE-FP, School of Psychology, University of Kent, Canterbury, UK.

Gannon, T. A., Ó Ciardha, C., Barnoux, M. F. L., Tyler, N., Mozova, K., & Alleyne, E. K. A. (2013). Maleimprisoned firesetters have different characteristics than other imprisoned offenders andrequire specialist treatment. Psychiatry: Interpersonal and Biological Processes, 76, 349–364.doi:10.1521/psyc.2013.76.4.349

Gannon, T. A., Ó Ciardha, C., Doley, R. M., & Alleyne, E. (2012). The multi-trajectory theory of adult fire-setting. Aggression and Violent Behavior, 17, 107–121. doi:10.1016/j.avb.2011.08.001

Gannon, T. A., & Pina, A. (2010). Firesetting: Psychopathology, theory and treatment. Aggression andViolent Behaviour, 15, 224–238. doi:10.1016/j.avb.2010.01.001

Geller, J. L. (1992). Communicative arson. Hospital and Community Psychiatry, 43, 76–77.George, D., & Mallery, P. (2003). SPSS for windows step by step: A simple guide and reference, 11.0

update (4th ed.). Boston, MA: Allyn & Bacon.Gov.UK. (2015).What happens if a child under 10 breaks the law? Retrieved from https://www.gov.uk/

child-under-10-breaks-lawHarmon, R. B., Rosner, R., & Wiederlight, M. (1985). Women and arson: A demographic study. Journal

of Forensic Sciences, 30(2), 467–477.Hopley, A. A. B., & Nicki, R. M. (2010). Predictive factors of excessive online poker playing.

CyberPsychology, Behavior & Social Networking, 13(4), 379–385. doi:10.1089/cyber.2009.0223Hurley, W., & Monahan, T. M. (1969). Arson: The criminal and the crime. British Journal of Criminology,

9(1), 4–21.Icove, D. J., & Estepp, M. H. (1987). Motive-based offender profiles of arson and fire-related crimes, FBI

Law Enforcement Bulletin. Washington, DC: U.S. Department of Justice.Inciardi, J. (1970). The adult firesetter. Criminology, 8, 145–155.Jackson, H., Glass, C., & Hope, S. (1987). A functional analysis of recidivistic arson. British Journal of

Clinical Psychology, 26, 175–185.Jayaraman, A., & Frazer, J. (2006). Arson: A growing inferno.Medicine, Science and the Law, 46(4), 295–300.Jenerette, C., & Dixon, J. (2010). Developing a short form of the simple Rathus assertiveness schedule

using a sample of adults with sickle cell disease. Journal of Transcultural Nursing, 21(4), 314–324.doi:10.1177/1043659609360712

PSYCHOLOGY, CRIME & LAW 19

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

Jones, J. P., Thomas-Peter, B. A., & Trout, A. (1999). Brief report. Normative data for the Novaco AngerScale from a non-clinical sample and implications for clinical use. British Journal of ClinicalPsychology, 38, 417–424.

Justice Statistics Analytical Services – Ministry of Justice. (2015, May). Information gained from afreedom of information act request to the ministry of justice. London: Author (Ref: 237–15 FoI 97457).

Kazdin, A. E., & Kolko, D. L. (1986). Parent psychopathology and family functioning among childhoodfiresetters. Journal of Abnormal Child Psychology, 14(2), 315–329.

Kennedy, P., Vale, E., Khan, S., & McAnaney, A. (2006). Factors predicting recidivism in child and ado-lescent fire-setters: A systematic review of the literature. The Journal of Forensic Psychiatry &Psychology, 17, 151–164. doi:10.1080/1478994050044150

Kolko, D. J. (2001). Efficacy of cognitive-behavioural treatment and fire safety education for firesettingchildren: Initial and follow-up outcomes. Journal of Child Psychology and Psychiatry, 42(3), 359–369.

Kolko, D. J. (Ed.). (2002). Handbook on firesetting in children and youth. San Diego, CA: Academic Press.Koson, D. F., & Dvoskin, J. (1982). Arson: A diagnostic study. Bulletin of American Academy of Psychiatry

and the Law, 10(1), 39–49.Lewis, M. D., & Yarnell, H. (1951). Pathological firesetting (pyromania). Nervous and Mental Disease

Monographs, 82, 30–37.Macht, L. B., & Mack, J. E. (1968). The fire-setter syndrome. Psychiatry, 31, 277–288.MacKay, S., Henderson, J., Del Bove, G., Marton, P., Warling, D., & Root, C. (2006). Fire interest and anti-

sociality as risk factors in the severity and persistence of juvenile firesetting. Journal of theAmerican Academy of Child and Adolescent Psychiatry, 45(9), 1077–1084. doi:10.1097/01.chi.0000227881.50404.ca

McCardle, S., Lambie, I., & Barker-Collo, S. (2004). Adolescent firesetting: A New Zealand case-controlledstudy of risk factors for adolescent firesetting. New Zealand Fire Service Commission ResearchReport Number 46.

McKerracher, D., & Dacre, A. (1966). A study of arsonists in a special security hospital. British Journal ofPsychiatry, 112, 1151–1154.

Mills, J. F., & Kroner, D. G. (1999). Measures of criminal attitudes and associates: User guide(Unpublished instrument and user guide).

Mills, J. F., Kroner, D. G., & Forth, A. E. (2002). Measures of Criminal Attitudes and Associates (MCAA):Development, factor structure, reliability, and validity. Assessment, 9(3), 240–253.

Molnar, G., Keitner, L., & Harwood, B. T. (1984). A comparison of partner and solo arsonists. Journal ofForensic Sciences, 29(2), 574–583.

Muckley, A. (1997). Firesetting: Addressing offending behaviour. A resource and training manual. Redcarand Cleveland: Redcar and Cleveland Psychological Service.

Muller, D. A. (2008). Offending and reoffending patterns of arsonists and bushfire arsonists in NewSouth Wales. Trends and issues in crime and criminal justice. Australian Institute of Criminology,348, 1–6.

Murphy, G. H., & Clare, I. C. H. (1996). Analysis of motivation in people with mild learning disabilities(mental handicap) who set fires. Psychology, Crime & Law, 2(3), 153–164. doi:10.1080/10683169608409774

Noblett, S., & Nelson, B. (2001). A psychosocial approach to arson – A case controlled study of femaleoffenders. Medicine Science and the Law, 41(4), 325–330.

Novaco, R. W. (2003). The Novaco Anger Scale and provocation inventory: NAS-PI. Los Angeles, CA:Western Psychological Services.

Nowicki, S., Jr. (1976). Adult Nowicki-Strickland internal-external locus of control scale. Test manualavailable from S. Nowicki, Jr., Department of Psychology, Emory University, Atlanta, GA 30322.

Ó Ciardha, C., & Gannon, T. A. (2012). The implicit theories of firesetters: A preliminary conceptualiz-ation. Aggression and Violent Behavior, 17, 122–128. doi:10.1016/j.avb.2011.12.001

O’Sullivan, G. H., & Kelleher, M. J. (1987). A study of firesetters in the South-West of Ireland. BritishJournal of Psychiatry, 151, 818–823. doi:10.1192/bjp.151.6.818

Paulhus, D. L. (1984). Two-component models of socially desirable responding. Journal of Personalityand Social Psychology, 46, 598–609. doi:10.1037/0022-3514.46.3.598

20 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

Paulhus, D. L. (1988). Assessing self deception and impression management in self reports: The balancedinventory of desirable responding (Unpublished manual). University of British Columbia, Vancouver.

Perrin-Wallqvist, R., Archer, T., & Norlander, T. (2004). Adolescents’ fire-setting awareness underboredom: Relation to personality variables. Psychological Reports, 94(3), 863–871.

Pettiway, L. E. (1987). Arson for revenge: The role of environmental situation, age, sex, and race.Journal of Quantitative Criminology, 3(2), 169–184.

Prins, H. (1994). Fire-raising. Its motivation and management. London: Routledge.Räsänen, P., Hakko, H., & Väisänen, E. (1995). The mental state of arsonists as determined by forensic

psychiatric examinations. The Bulletin of the American Academy of Psychiatry and the Law, 23(4),547–553.

Rautaheimo, J. (1989). The making of an arsonist. Fire Prevention, 223, 30–34.Rice, M. E., & Chaplin, T. C. (1979). Social skills training for hospitalized male arsonists. Journal

of Behavior Therapy and Experimental Psychiatry, 10(2), 105–108. doi:10.1016/0005-7916(79)90083-1

Rix, K. J. B. (1994). A psychiatric study of adult arsonists. Medicine Science and the Law, 34(1),21–34.

Root, C., MacKay, S., Henderson, J., Del Bove, G., & Warling, D. (2008). The link between maltreatmentand juvenile firesetting: Correlates and underlying mechanisms. Child Abuse & Neglect, 32,161–176. doi:10.1016/j.chiabu.2007.07.004

Russell, D., Peplau, L. A., & Cutrona, C. E. (1980). The revised UCLA loneliness scale: Concurrent anddiscriminant validity evidence. Journal of Personality and Social Psychology, 39(3), 472–480.doi:10.1037//0022-3514.39.3.472

Sapp, A. D., Huff, T. G., Gary, G. P., & Icove, D. J. (1999). Serial arson and fire-related crime factors. In V.B. Van Hasselt & M. Hersen (Eds.), Handbook of psychological approaches with violent offenders:Contemporary strategies and issues (pp. 397–406). New York: Kluwer Academic/Plenum.

Sapsford, R. J., Banks, C., & Smith, D. D. (1978). Arsonists in prison.Medicine, Science and the Law, 18(4),247–254. doi:10.1177/002580247801800405

Saunders, E. B., & Awad, G. A. (1991). Adolescent female firesetters. Canadian Journal of Psychiatry, 36(6), 401–404.

Sax, L. J., Gilmartin, S. K., Lee, J. J., & Hagedorn, L. S. (2008). Using web surveys to reach communitycollege students: An analysis of response rates and response bias. Community College Journal ofResearch and Practice, 32(9), 712–729. doi:10.1080/10668920802000423

Smith, K., Taylor, P., & Elkin, M. (2013). Crimes detected in England and Wales 2012/13 (2nd ed.).London: National Statistics.

Swaffer, T., & Hollin, C. R. (1995). Adolescent firesetting: Why do they say they do it? Journal ofAdolescence, 18, 619–623. doi:10.1006/jado.1995.1043

Tennent, T. G., McQuaid, A., Loughnane, T., & Hands, A. J. (1971). Female arsonists. British Journal ofPsychiatry, 119, 497–502.

Tyler, N., & Gannon, T. A. (2012). Explanations of firesetting in mentally disordered offenders: A reviewof the literature. Psychiatry, 75(2), 150–166. doi:10.1521/psyc.2012.75.2.150

Tyler, N., Gannon, T. A., Lockerbie, L., King, T., Dickens, G. L., & De Burca, C. (2014). A firesetting offencechain for mentally disordered offenders. Criminal Justice and Behaviour, 41(4), 512–530. doi:10.1177/0093854813510911

Underwood, D., Kim, H., & Matier, M. (2000, May 21–24). To mail or to web: Comparisons of surveyresponse rates and respondent characteristics. Paper presented at the 40th annual forum of theAssociation for Institutional Research, Cincinnati, OH.

Vassar, M., & Crosby, J. W. (2008). A reliability generalization study of coefficient alpha for the UCLAloneliness scale. Journal of Personality Assessment, 90(6), 601–607.

Vaughn, M. G., Fu, Q., Delisi, M., Wright, J. P., Beaver, K. M., Perron, B. E., & Howard, M. O. (2010).Prevalence and correlates of fire-setting in the United States: Results from the NationalEpidemiological survey on alcohol and related conditions. Comprehensive Psychiatry, 51,217–223. doi:10.1016/j.comppsych.2009.06.002

PSYCHOLOGY, CRIME & LAW 21

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

Vodanovich, S. J., Wallace, J. C., & Kass, S. J. (2005). A confirmatory approach to the factor structure ofthe Boredom Proneness Scale: Evidence for a two-factor short form. Journal of PersonalityAssessment, 85(3), 295–303. doi:10.1207/s15327752jpa8503_05

Wachi, T., Watanabe, K., Yokota, K., Suzuki, M., Hoshino, M., Sato, A., & Fujita, G. (2007). Offender andcrime characteristics of female serial arsonists in Japan. Journal of Investigative Psychology andOffender Profiling, 4, 29–52. doi:10.1002/jip.57

Watt, B. D., Geritz, K., Hasan, T., Harden, S., & Doley, R. (2015). Prevalence and correlates of firesettingbehaviours among offending and non-offending youth. Legal and Criminological Psychology, 20,19–36. doi:10.1111/lcrp.12062

Williams, D. (2005). Understanding the arsonist: From assessment to confession. Tucson, AZ: Lawyersand Judges Publishing Company.

Appendix 1. Items from the FSS

The following items were presented using a 7-point Likert scale (1 = not at all like me, 7 = verystrongly like me).

Fire Interest items

I like to watch and feel fireI get excited thinking about fireI like watching fireI like watching fire being extinguishedI like to feel the heat from fireI am fascinated by fireI have a strong interest in fireI am attracted to fireFire equipment paraphernalia interests meI find fire intriguing

Antisocial Behaviour items

At school I would often truantI like to engage in acts that are dangerousI have a behavioural problemI have intended to cause harm with my behaviourI am a rule breakerI like to engage in acts that are excitingI like to wind people upI care what other people think of meI like to engage in acts to annoy other peopleI have physically threatened another person

Appendix 2. FPS Vignettes

The following items were introduced with the next six questions involve reading a short story, andanswering four questions relating to the story.

Billie is a 15-year-old who had spent the weekend being bored. Billie decided to go to the localwreck to see if anyone wanted to hang out. There were already a few people there just hangingaround and chatting. One of them lit a cigarette. The sight of the flame shooting out of thelighter gave Billie an idea. Billie decided to set a rubbish bin alight. Billie lit a piece of rubbish and

22 E. R. BARROWCLIFFE AND T. A. GANNON

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

dropped it into the bin. The rest of the rubbish burned and the bin began to melt whilst Billie and thegroup carried on chatting and hanging out.

Tony felt constrained by life, conforming to the rules and regulations of society but in the countryTony felt free and relaxed. Nature appealed to Tony because it is free and natural. One quiet Sundayevening Tony decided to light a twig on fire. Tony watched as the flames were also free to flicker andmove as they pleased. From the burning twig, Tony then lit a pile of dried leaves and watched andlistened as the leaves crackled in the flames.

Hillary had finished sorting through the paperwork and had accumulated a large pile of oldpapers. Hillary took the old papers to the bottom of the garden and put them in a pile. Hillarythen lit the corners of a few of the papers at the bottom of the pile. Hillary stood back andwatched as the flames slowly crept up the side of the stack of papers. Hillary watched as theflames danced about freely in the breeze engulfing the whole stack of papers until eventually theold pile of papers were reduced to a pile of ashes.

Jo and the other locals would often dare each other to play pranks on the adults in the street. Theneighbourhood was fairly posh and most people lived in large gated properties with big gardens.Some people had electric gates whilst others had picket fences but most people had letter and news-paper boxes attached to either their fence or gate. One day whilst Jo was delivering papers it wasagreed that when the paper was put into the newspaper box it would be set alight. So Jo lit thecorner of the paper and put it into the newspaper box and then carried on with the rest of thepaper round.

Terry had always had an interest in fire and became excited when thinking about fire. Often whenalone either at work or at home Terry would light matches. Terry watched as the intensity and thecolour of the flame changed as more of the match began to burn. As the flame began to die out butbefore totally extinguished Terry lit another match from the original flame. Terry was fascinated bythe falling trail of ash left behind by the burning match and by the intensity of the heat from one littleflame.

Sammy and the others in the group were very mischievous. They spent most of their weekendscreating some sort of graffiti on the local bus station walls. One weekend they decided to reduce theproblem of old bus tickets littering the floor by setting fire to them. This then progressed to lightingthe corners of posters hanging on the walls and watching them crinkle up and fall off the walls creat-ing little piles of ashes.

PSYCHOLOGY, CRIME & LAW 23

Dow

nloa

ded

by [

Uni

vers

ity o

f K

ent]

at 2

3:49

23

Nov

embe

r 20

15

The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.